The lipopolysaccharides (LPS) of Gram-negative bacteria have complex structures and multiple functions. These bacterial surface molecules are very important in interactions between bacteria and their environment, including beneficial and destructive interactions with animal and plant hosts. The proposed work deals with how the outer portion, the O antigen, of the LPS is synthesized. There are two well-differentiated mechanisms of O-antigen synthesis. The LPS of Rhizobium etli CE3 is one of the better characterized examples (structurally and biologically) in which the O-antigen portion of the LPS is synthesized by a mechanism dependent on an ABC transporter encoded by wzm and wzt genes. It fits into a relatively under-studied subgroup of such LPSs in which the O antigen is a heteropolysaccharide, rather than a homopolysaccharide. The importance of its O-antigen in the symbiosis with Phaseolus vulgaris (bean) has been established, the structure of the entire LPS is almost completely solved, most of the steps in the synthesis of the lipid A and core portion of the molecule have been demonstrated in vitro, and a 30 kb cluster of genes that specify most of O-antigen biosynthesis has been sequenced. Moreover, the nucleotide sequence of the entire genome has been determined. In the proposed work, basic features of the biosynthesis of this O antigen would be determined. It would test a set of hypotheses that predict the order in which the transferases add the sugar residues and which genes encode which transferases. Most of the proposed work is biochemical. One approach that already is well-underway is to to deduce from truncated LPSs the steps at which different mutants are blocked. It has added to the hypothesis by matching genes with specific predicted glycosyltransferase activities. Another set of approaches involve examination of reactions in vitro with cell extracts from the various mutants predicted to have lesions in glycosyltransferases. In particular the work seeks to define the enzymes and products of the first reaction in the synthesis of the O antigen. In addition to this biochemical work, there will be additional genetic analysis to finish the nearly complete accounting of which genes are required in O-antigen synthesis in this bacterium and which functions they might have. These latter experiments will generate specific, nonpolar mutations in genes that still have not been mutated in two genetic loci that we have strong evidence are devoted to O-antigen synthesis. In addition two types of targeted genetic manipulations will address two interesting questions that may be related to the extreme uniformity of length of this O antigen. PUBLIC HEALTH RELEVANCE: With a model bacterium that does not cause medical problems, this project examines the fundamental basis of a property of health-beneficial and pathogenic bacteria that is crucial to their survival, especially in the hostile confines of the mammalian circulatory fluids. It also is important in other interactions with animal and plant hosts. This property is a type of carbohydrate chain on the surfaces of many bacteria. The question addressed in this work is how this chain is synthesized.